# Why isn't the signal reaching ground? - Simple FET Question

I have a p-ch mosfet with the gate being driven by a microcontroller at 1.2Khz 50% duty cycle between 3v and 0v. The mosfet threshold voltage is achieved.

But I don't understand why the waveform is not reaching 0V. The scope is probed between the source (Va) and the drain (GND) of the mosfet. I would assume during each "ON" cycle of the FET the waveform would reach 0V.

Additionally, why is the waveform not between 10V and 0V? Instead its between 5.8V and about 3V.

• P-channel FET are used on the high side, not the low side. Your VGS probably doesn't reach the threshold for full conduction. Commented Nov 9, 2022 at 20:38
• You are using the MOSFET in a source-follower configuration and it looks like you are applying the signal for a logic-level N-channel MOSFET.
– vir
Commented Nov 9, 2022 at 20:39
• You've drawn the bulk diode incorrectly <-- clue! Commented Nov 9, 2022 at 21:56

The voltage gain of a common drain/source follower setup is ~1, your results are expected behavior.

To get the behavior you want, you'll have to use a logic level N-channel MOSFET. The device you have has a threshold voltage of -2 to -4V. Besides being the wrong polarity, you'll only be able to pass a few mA at best with the 3V input you're giving it, and likely an order of magnitude less. Replace the power P-channel MOSFET with something like a TN0702 that is designed to have a reasonable RDSon with 3V of gate drive.

Realized the issue. The resistor needs to be flipped to the drain side of the p-fet, between the drain and ground. This allows the threshold voltage to be achieved when the mosfet is activated.

In the current circuit, once the mosfet is activated the source is immediately connected to ground thereby violating the threshold requirement and causing the mosfet to open. This cycle repeats endlessly.

Additionally, my driving source should be between 0V and -4V as some have pointed out. But I don't have that capability so my intent was to achieve the threshold during the negative duty cycle of the period.

The waveform is with Vcc set to 5V instead of 10V. I would need to pull the gate to 10V with a resistor but because I am driving the gate with a micro-controller I do not want to have 10V interact with the GPIO and I don't have a diode.

Thank you all for the input.